headers
Juan Ruiz | 3 Mar 2008 01:06

Re: [Sigia-l] Automatic personaliztion of the look-and-feel

Hi Andres,

>From my experience working in this field in Australia, here is what I can contribute

>>>- What is your opinion of the ability to personalize the interface in such a way?
For this type of forms, which has a pre-determined goal, the customization of forms might not add much
benefit to it. The simplicity and efficiency of the form, in the other hand, will improve the user experience.

>>>>- Would you think it would increase form completion?
>From various usability studies, when system responds with major visual changes (in your example, with
the visual design), users tend to try different combinations to see what other designs the site will
offer. This behavior is based on curiosity rather than need.

>>>>- Do you think it would improve brand perception / segment alignment?
This sort of feature will display the technical capabilities of the company rather than the brand
perception / segment alignment

>>>>- Or on the other hand, do you believe skinning does not help?
I would focus on the usability and efficiency of the form. Remove questions that are not necessary, and have
the system autocomplete or suggest possible answers.

Hope that helps,
Regards,
-Juan Ruiz

------------
IA Summit 2008: "Experiencing Information" 
April 10-14, 2008, Miami, Florida

-----
(Continue reading)

Permalink | Reply | 2 comments |
headers
Jessica Enders | 3 Mar 2008 08:57
Picon

Re: [Sigia-l] Automatic personaliztion of the look-and-feel

I would have to agree with everything Juan said.

When people are presented with a form, they want to complete it and  
get on with their life as quickly and easily as possible. I don't  
have any references to hand, but I'm sure I've read research that  
also shows users they are heavily "field focused" and often won't  
notice other things (which includes customisation). This has  
certainly been my experience from user testing.

Good luck

Jessica Enders
Director
Formulate Information Design
----------------------------------------
http://formulate.com.au
----------------------------------------
Phone: (02) 6116 8765
Fax: (02) 8456 5916
PO Box 5108
Braddon ACT 2612
----------------------------------------

On 03/03/2008, at 11:06 AM, Juan Ruiz wrote:

> Hi Andres,
>
>> From my experience working in this field in Australia, here is  
>> what I can contribute
>
(Continue reading)

Permalink | Reply | 1 comment |
headers
Sarah Dilling | 3 Mar 2008 17:53
Picon

[Sigia-l] tagging software

Hi there,

I have a pile of things (internal mailing lists, if you're curious).
I'd like to tag them and have software generate a tag cloud for me.
Doesn't need to be web accessible or anything, just my curiosity and
perhaps a jumping off point.

Does anyone know of a free-standing piece of tag software that does
this? Or do I have to abuse del.icio.us or write my own?

SarahD
------------
IA Summit 2008: "Experiencing Information" 
April 10-14, 2008, Miami, Florida

-----
When replying, please *trim your post* as much as possible.
*Plain text, please; NO Attachments

Searchable Archive at http://www.info-arch.org/lists/sigia-l/
________________________________________
Sigia-l mailing list -- post to: Sigia-l <at> asis.org
Changes to subscription: http://mail.asis.org/mailman/listinfo/sigia-l
Permalink | Reply | 10 comments |
headers
W Evans | 3 Mar 2008 18:19
Picon

Re: [Sigia-l] tagging software

Isn't the whole point behind the CogSci of tagging of content to generate
folksonomies is that it's done by many people and the content get's tag
attributes organically through the wisdom of crowds. If a machine is going
to tag something - you might as well employ full blown natural language
processing to pull out entities and relationships from content. But this is
not, nor can it be, the point behind tagging.

On Mon, Mar 3, 2008 at 11:53 AM, Sarah Dilling <sarahd <at> gmail.com> wrote:

> Hi there,
>
> I have a pile of things (internal mailing lists, if you're curious).
> I'd like to tag them and have software generate a tag cloud for me.
> Doesn't need to be web accessible or anything, just my curiosity and
> perhaps a jumping off point.
>
> Does anyone know of a free-standing piece of tag software that does
> this? Or do I have to abuse del.icio.us or write my own?
>
> SarahD
> ------------
> IA Summit 2008: "Experiencing Information"
> April 10-14, 2008, Miami, Florida
>
> -----
> When replying, please *trim your post* as much as possible.
> *Plain text, please; NO Attachments
>
> Searchable Archive at http://www.info-arch.org/lists/sigia-l/
> ________________________________________
(Continue reading)

Permalink | Reply | 9 comments |
headers
Frank Shepard | 3 Mar 2008 18:34
Picon

Re: [Sigia-l] tagging software

"If a machine is going to tag something - you might as well employ
full blown natural language
processing to pull out entities and relationships from content. But this is
not, nor can it be, the point behind tagging."

Why not? if the context is a single user and her data, why not allow a
machine to generate tags? What's the objection to that?

Frank

On Mon, Mar 3, 2008 at 12:19 PM, W Evans <wkevans4 <at> gmail.com> wrote:
> Isn't the whole point behind the CogSci of tagging of content to generate
>  folksonomies is that it's done by many people and the content get's tag
>  attributes organically through the wisdom of crowds. If a machine is going
>  to tag something - you might as well employ full blown natural language
>  processing to pull out entities and relationships from content. But this is
>  not, nor can it be, the point behind tagging.
>
>
>
>  On Mon, Mar 3, 2008 at 11:53 AM, Sarah Dilling <sarahd <at> gmail.com> wrote:
>
>  > Hi there,
>  >
>  > I have a pile of things (internal mailing lists, if you're curious).
>  > I'd like to tag them and have software generate a tag cloud for me.
>  > Doesn't need to be web accessible or anything, just my curiosity and
>  > perhaps a jumping off point.
>  >
>  > Does anyone know of a free-standing piece of tag software that does
(Continue reading)

Permalink | Reply | 8 comments |
headers
W Evans | 3 Mar 2008 18:46
Picon

Re: [Sigia-l] tagging software

Because then it's not tagging, it's generating a custom taxonomy without the
hierarchy.

On Mon, Mar 3, 2008 at 12:34 PM, Frank Shepard <fgshepard <at> gmail.com> wrote:

> "If a machine is going to tag something - you might as well employ
> full blown natural language
> processing to pull out entities and relationships from content. But this
> is
> not, nor can it be, the point behind tagging."
>
> Why not? if the context is a single user and her data, why not allow a
> machine to generate tags? What's the objection to that?
>
> Frank
>
> On Mon, Mar 3, 2008 at 12:19 PM, W Evans <wkevans4 <at> gmail.com> wrote:
> > Isn't the whole point behind the CogSci of tagging of content to
> generate
> >  folksonomies is that it's done by many people and the content get's tag
> >  attributes organically through the wisdom of crowds. If a machine is
> going
> >  to tag something - you might as well employ full blown natural language
> >  processing to pull out entities and relationships from content. But
> this is
> >  not, nor can it be, the point behind tagging.
> >
> >
> >
> >  On Mon, Mar 3, 2008 at 11:53 AM, Sarah Dilling <sarahd <at> gmail.com>
(Continue reading)

Permalink | Reply | 6 comments |
headers
W Evans | 3 Mar 2008 18:50
Picon

Re: [Sigia-l] tagging software

I also think one of the biggest benefits of tagging is information discovery
and findability. This means surfing down tag clouds of artifacts other
people tagged. The upfront cost of tagging approaches zero - especially when
distributed. If you are both the tagger and the consumer of the tag....
wouldn't you just create an organization scheme you can stick to?

On Mon, Mar 3, 2008 at 12:46 PM, W Evans <wkevans4 <at> gmail.com> wrote:

> Because then it's not tagging, it's generating a custom taxonomy without
> the hierarchy.
>
> On Mon, Mar 3, 2008 at 12:34 PM, Frank Shepard <fgshepard <at> gmail.com>
> wrote:
>
> > "If a machine is going to tag something - you might as well employ
> > full blown natural language
> > processing to pull out entities and relationships from content. But this
> > is
> > not, nor can it be, the point behind tagging."
> >
> > Why not? if the context is a single user and her data, why not allow a
> > machine to generate tags? What's the objection to that?
> >
> > Frank
> >
> > On Mon, Mar 3, 2008 at 12:19 PM, W Evans <wkevans4 <at> gmail.com> wrote:
> > > Isn't the whole point behind the CogSci of tagging of content to
> > generate
> > >  folksonomies is that it's done by many people and the content get's
> > tag
(Continue reading)

Permalink | Reply | 5 comments |
headers
Frank Shepard | 3 Mar 2008 19:00
Picon

Re: [Sigia-l] tagging software

but, in this case, the user isn't the tagger. the computer is. crowds
don't have a monopoly on "wisdom" ...

On Mon, Mar 3, 2008 at 12:50 PM, W Evans <wkevans4 <at> gmail.com> wrote:
> I also think one of the biggest benefits of tagging is information discovery
> and findability. This means surfing down tag clouds of artifacts other
> people tagged. The upfront cost of tagging approaches zero - especially when
> distributed. If you are both the tagger and the consumer of the tag....
> wouldn't you just create an organization scheme you can stick to?
>
>
>
> On Mon, Mar 3, 2008 at 12:46 PM, W Evans <wkevans4 <at> gmail.com> wrote:
> > Because then it's not tagging, it's generating a custom taxonomy without
> the hierarchy.
> >
> >
> >
> >
> >
> > On Mon, Mar 3, 2008 at 12:34 PM, Frank Shepard <fgshepard <at> gmail.com>
> wrote:
> >
> > >
> > > "If a machine is going to tag something - you might as well employ
> > > full blown natural language
> > > processing to pull out entities and relationships from content. But this
> is
> > > not, nor can it be, the point behind tagging."
> > >
(Continue reading)

Permalink | Reply | 4 comments |
headers
W Evans | 3 Mar 2008 19:07
Picon

Re: [Sigia-l] tagging software

There are certainly a couple (mostly in very very early stage development),
that will do "tagging" of a person's content by the machine once it's in a
machine readable format. But it's not called tagging, it's called semantic
analysis - and it's not quite there yet (under very very controlled academic
circumstances where all data is normalized beforehand, I know one NLP
researcher at MIT able to get matching accuracy of machine tags to 85+%. In
the real world - it's around 50% given perfect grammar and no slang or
idioms

On Mon, Mar 3, 2008 at 1:00 PM, Frank Shepard <fgshepard <at> gmail.com> wrote:

> but, in this case, the user isn't the tagger. the computer is. crowds
> don't have a monopoly on "wisdom" ...
>
> On Mon, Mar 3, 2008 at 12:50 PM, W Evans <wkevans4 <at> gmail.com> wrote:
> > I also think one of the biggest benefits of tagging is information
> discovery
> > and findability. This means surfing down tag clouds of artifacts other
> > people tagged. The upfront cost of tagging approaches zero - especially
> when
> > distributed. If you are both the tagger and the consumer of the tag....
> > wouldn't you just create an organization scheme you can stick to?
> >
> >
> >
> > On Mon, Mar 3, 2008 at 12:46 PM, W Evans <wkevans4 <at> gmail.com> wrote:
> > > Because then it's not tagging, it's generating a custom taxonomy
> without
> > the hierarchy.
> > >
(Continue reading)

Permalink | Reply | 3 comments |
headers
W Evans | 3 Mar 2008 19:08
Picon

Re: [Sigia-l] tagging software

Here is the paper:

Pawan Deshpande. Thesis Supervisor: Regina
Barzilay.<http://people.csail.mit.edu/regina>
Decoding Algorithms for Complex Natural Language
Tasks<http://people.csail.mit.edu/pawand/t/main.pdf>.
M. Eng. Thesis <http://people.csail.mit.edu/pawand/t/main.pdf>.

On Mon, Mar 3, 2008 at 1:07 PM, W Evans <wkevans4 <at> gmail.com> wrote:

> There are certainly a couple (mostly in very very early stage
> development), that will do "tagging" of a person's content by the machine
> once it's in a machine readable format. But it's not called tagging, it's
> called semantic analysis - and it's not quite there yet (under very very
> controlled academic circumstances where all data is normalized beforehand, I
> know one NLP researcher at MIT able to get matching accuracy of machine tags
> to 85+%. In the real world - it's around 50% given perfect grammar and no
> slang or idioms
>
>
> On Mon, Mar 3, 2008 at 1:00 PM, Frank Shepard <fgshepard <at> gmail.com> wrote:
>
> > but, in this case, the user isn't the tagger. the computer is. crowds
> > don't have a monopoly on "wisdom" ...
> >
> > On Mon, Mar 3, 2008 at 12:50 PM, W Evans <wkevans4 <at> gmail.com> wrote:
> > > I also think one of the biggest benefits of tagging is information
> > discovery
> > > and findability. This means surfing down tag clouds of artifacts other
> > > people tagged. The upfront cost of tagging approaches zero -
(Continue reading)

Permalink | Reply | 1 comment |

Gmane